As the percentage of our population over 65 increases, concern about cognitive changes in the healthy elderly grows. All higher order cognition (e.g., remembering autobiographical events, reasoning and problem solving) depends critically on two broad classes of memory function, working memory (WM) and long-term memory (LTM). Aging is associated with deficits in both WM and LTM and such deficits have an impact on the cognitive functioning of older adults in all areas of their lives. WM refers to the information from perception and long-term memory that is currently active, and the set of processes that maintain and manipulate this active information. It allows us to keep something in mind after the initiating stimulus disappears, to make connections and comparisons between events, and to sustain goal-directed behavior. Furthermore, WM processes are "encoding" processes for information that will be available later in LTM as part of the cumulative record of our past experiences. Neuronal studies with monkeys, studies of patients with brain damage, and recent human neuroimaging studies point to the prefrontal cortex (PFC) as critical for WM and LTM. In addition, there is evidence of age-related neuropathology that may disproportionately affect PFC, suggesting that changes in PFC function may, in part, underlie memory dysfunction in older adults. The proposed research combines a component process approach to memory with functional magnetic resonance imaging (fMRI) techniques in order to characterize more precisely, cognitively and neurally, memory deficits associated with aging. The proposed project has four specific aims: (1) To further clarify the component processes of memory and to identify their neural bases. (2) To identify and specify the neural bases of those component processes that are (and are not) affected by aging. (3) To clarify the conditions under which age-related compensatory brain activity is found and the nature of such activity. (4) To explore the relations between component processes engaged during WM (or encoding) and long-term memory and the implications for long-term memory of age-related changes in these component processes.